Tracking the Impact of Wildfire on Urban Air Pollution: A Satellite-Based Assessment of 2025 Los Angeles Wildfire

1. Introduction

Wildfires have become increasingly frequent, widespread, and destructive across the United States (US), particularly in the western regions, where prolonged droughts, elevated temperatures, and land-use changes have amplified fire risk. Over the past four decades, the annual area burned has more than quadrupled, with wildland-urban interface (WUI) regions facing elevated vulnerability due to population growth and vegetation encroachment [1,2]. These fires pose dual threats i.e., destroying natural ecosystems and infrastructure, while also emitting massive quantities of air pollutants that compromise atmospheric composition and public health across local and regional scales [3]. The atmospheric impacts of wildfires, especially in urban and peri-urban regions, extend beyond the fire perimeter, as emissions from burning biomass release fine particulate matter (PM_2.5), nitrogen oxides (NO_x), carbon monoxide (CO), and volatile organic compounds (VOCs), many of which are precursors to ozone formation [4,5,6].

In January 2025, a series of rapidly spreading wildfires erupted across Southern California, fueled by unseasonably warm and dry offshore Santa Ana winds [7,8]. The most intense activity occurred in the greater Los Angeles region, where multiple fires including the Griffith Fire and San Bernardino blaze collectively scorched over 45,000 acres and forced the evacuation of tens of thousands of residents. These fires, striking during what is typically a low-risk season, not only challenged emergency responders but also triggered an abrupt and hazardous deterioration in urban air quality [8].

Numerous wildfire events in the US have demonstrated the severe degradation of air quality and associated health impacts on both local and regional populations. For example, the Camp Fire (2018) in Northern California, which destroyed the town of Paradise, led to daily PM_2.5 concentrations in Sacramento and San Francisco that exceeded 200 µg/m³ which classified as “very unhealthy” to “hazardous” by the U.S. EPA [9]. The smoke plume extended over 1,500 km, affecting air quality in neighboring states [10]. Similarly, the Woolsey Fire (2018) in Southern California impacted the Los Angeles Basin, elevating levels of fine particulate matter and O₃ in an already polluted urban environment [11]. In 2020, the August Complex Fire, the largest in California’s history, along with simultaneous fires in Oregon and Washington, resulted in the highest AQI values ever recorded in Portland, Seattle, and San Francisco, forcing widespread closures of schools, businesses, and transit systems [12].

Beyond the West Coast, smoke transport from western wildfires has repeatedly affected regions in the central and eastern U.S. In July 2021, wildfires in British Columbia and the Pacific Northwest produced smoke that blanketed the Midwest and Northeastern U.S., degrading visibility in New York City and increasing PM_2.5 levels to over 50 µg/m³ which is well above the safe daily threshold [13]. During the same period, cities such as Chicago, Detroit, and Philadelphia issued air quality advisories, emphasizing the far-reaching impacts of wildfire emissions [14]. Moreover, the 2016 Great Smoky Mountains fires in Tennessee, which burned over 17,000 acres and led to 14 deaths, highlighted the growing wildfire risk in the eastern U.S., where vegetation and terrain differ substantially from the arid West but are increasingly susceptible under changing climate conditions [15].

Exposure to wildfire-related pollution has been robustly linked to increased hospital admissions for asthma, bronchitis, ischemic heart disease, and premature mortality, especially among children, the elderly, and individuals with pre-existing health conditions [16,17,18]. Furthermore, the economic burden of wildfire smoke such as lost productivity, elevated healthcare costs, and school disruptions can be regarded as direct damages of the fires themselves. Despite these severe implications, continuous and high-resolution monitoring of wildfire-induced air quality changes is essential particularly in urban centers like Los Angeles. Ground-based monitoring stations, though accurate, are spatially sparse and often cannot resolve the plume evolution or detect pollution over mountainous and rural fire-affected zones [19]. Satellite remote sensing thus provides a valuable supplement, offering wide-area, temporally frequent observations of atmospheric pollutants. Instruments such as Sentinel-5P/TROPOMI (for NO₂, CO), MODIS and VIIRS (for Aerosol Optical Depth as a proxy for PM_2.5), and GOES geostationary satellites (for fire detection and smoke tracking) are essential for quantifying pollution transport and intensity during wildfire episodes [20,21].

This study investigates the air quality impacts of the January 2025 Southern California wildfires, focusing specifically on the Los Angeles metropolitan area. Using multi-sensor satellite observations, we examine how key atmospheric pollutants i.e., PM (via aerosol optical depth), nitrogen dioxide (NO₂), and carbon monoxide (CO)—varied temporally across three distinct phases of the wildfire: pre-fire, active fire, and post-fire periods. The primary objective of this research is to characterize the changes in urban air quality associated with the wildfire event and to identify the magnitude and temporal pattern of pollution anomalies. By tracking pollutant levels across these phases, this study contributes to a better understanding of the atmospheric consequences of urban-proximate wildfires and highlights the value of satellite-based remote sensing in evaluating air quality dynamics during extreme fire events.

2. Materials and Methods

2.1. Study Area and Event Description

This study aims to evaluate the temporal evolution of atmospheric pollutants—namely aerosol optical depth (AOD), carbon monoxide (CO), nitrogen dioxide (NO₂), and ozone (O₃)—before, during, and after the January 2025 Southern California wildfires, with a focus on the Los Angeles metropolitan area. These fires, driven by intense Santa Ana winds and anomalously dry winter conditions, resulted in the burning of over 45,000 acres between January 06 and January 30, 2025. To capture the full extent of air quality changes, the study period was divided into three phases: the pre-fire period (December 22, 2024–January 05, 2025), the active fire period (January 06–30), and the post-fire period (January 31–February 10). Research manuscripts reporting large datasets that are deposited in a publicly available database should specify where the data have been deposited and provide the relevant accession numbers. If the accession numbers have not yet been obtained at the time of submission, please state that they will be provided during review. They must be provided prior to publication.

2.2. Satellite Datasets

To assess these changes, we employed satellite-based remote sensing datasets that provide high-resolution, wide-area observations of atmospheric pollutants. AOD data were obtained from the Moderate Resolution Imaging Spectroradiometer (MODIS) instruments aboard both the Terra and Aqua satellites. Specifically, we used the MOD04_L2 and MYD04_L2 products, which provide Level 2 AOD retrievals at 550 nm with a spatial resolution of 10 km. These products are widely used as proxies for surface-level particulate matter (PM_2.5) and allow for robust spatial and temporal comparisons of aerosol loading associated with wildfire smoke.

Tropospheric CO and NO₂ column densities were retrieved from the Tropospheric Monitoring Instrument (TROPOMI) onboard the Sentinel-5 Precursor satellite. The S5P_L2_CO product provides daily measurements of CO at a spatial resolution of approximately 7 km × 7 km, enabling the detection of elevated CO concentrations from biomass burning and subsequent transport. Similarly, the S5P_L2_NO2 product delivers tropospheric NO₂ vertical column densities at high resolution, suitable for capturing combustion-driven pollution and urban NOₓ enhancements. In addition, total column O₃ concentrations were extracted from the S5P_L2_O3 product, offering insight into potential photochemical changes driven by wildfire-emitted precursors. While TROPOMI does not directly provide surface ozone, columnar anomalies can reflect regional-scale ozone perturbations during intense smoke episodes.

All satellite datasets were downloaded in NetCDF or HDF5 format and preprocessed using Python libraries such as xarray, numpy, and SatPy. Quality assurance flags provided within each product were applied to exclude retrievals with cloud contamination, low confidence scores, or sun glint interference. For uniformity, all variables were resampled onto a common 0.05° spatial grid covering Southern California. Daily mean values were computed for each pollutant, and composite maps were generated for each fire phase to evaluate spatial trends. Time series analyses were conducted to identify pollutant peaks and the lag in post-fire atmospheric recovery. Additionally, pollutant anomalies during the fire and post-fire phases were calculated relative to the pre-fire baseline to quantify the magnitude of change.

3. Results and Discussions

3.1. Descriptive Statistics of Air Pollutants

The analysis of satellite-derived air quality indicators reveals distinct temporal variations in pollutant concentrations across the three wildfire phases—pre-fire, active fire, and post-fire. For O₃, both the mean and maximum concentrations showed relatively stable values during the pre-fire period, with mean daily O₃ levels averaging around 0.121 mol/m² and maximum values around 0.128 mol/m². During the fire, a slight increase in mean O₃ was observed, rising to 0.126 mol/m², while post-fire values declined to pre-event levels (~0.123 mol/m²), indicating a transient enhancement.

Figure 1. Spatial distribution of mean O₃ column concentrations (mol/m²) over the Los Angeles region during the three wildfire phases: (a) pre-fire (January 8–14, 2025), (b) active fire (January 15–24, 2025), and (c) post-fire (January 25–February 1, 2025).

Figure 1. Spatial distribution of mean O₃ column concentrations (mol/m²) over the Los Angeles region during the three wildfire phases: (a) pre-fire (January 8–14, 2025), (b) active fire (January 15–24, 2025), and (c) post-fire (January 25–February 1, 2025).

In contrast, NO₂ exhibited sharper variations. The mean NO₂ concentration prior to the fire was relatively low (mean: 0.00015 mol/m²), consistent with typical wintertime urban background levels. During the wildfire, NO₂ levels increased substantially, with average values rising to 0.00020 mol/m², likely due to the combustion of nitrogen-rich vegetation and interaction with urban emissions. The post-fire phase showed a partial recovery, with mean concentrations declining to 0.00017 mol/m², though not fully returning to baseline.

Figure 2. Spatial distribution of mean NO₂ column concentrations (mol/m²) over the Los Angeles region during the three wildfire phases: (a) pre-fire (b) active fire, and (c) post-fire.

Figure 2. Spatial distribution of mean NO₂ column concentrations (mol/m²) over the Los Angeles region during the three wildfire phases: (a) pre-fire (b) active fire, and (c) post-fire.

CO displayed the most pronounced fire-induced enhancement. Pre-fire CO concentrations averaged 0.035 mol/m², whereas during the active fire period, mean levels rose significantly to 0.040 mol/m², accompanied by a peak daily value of 0.147 mol/m², nearly 4× higher than pre-fire maxima. This dramatic increase aligns with CO’s known role as a primary emission product from incomplete biomass combustion. After the fire, CO levels declined but remained slightly elevated (mean: 0.033 mol/m²) compared to the baseline.

Figure 3. Spatial distribution of mean CO column concentrations (mol/m²) over the Los Angeles region during the three wildfire phases: (a) pre-fire, (b) active fire, and (c) post-fire.

Figure 3. Spatial distribution of mean CO column concentrations (mol/m²) over the Los Angeles region during the three wildfire phases: (a) pre-fire, (b) active fire, and (c) post-fire.

3.2. Change Analysis from Pre-Fire Baseline

Figure 4 and Table 1 illustrates the temporal evolution of mean concentrations of O₃, NO₂, and CO across the pre-fire, active fire, and post-fire phases of the January 2025 Southern California wildfires. Each pollutant exhibited distinct temporal behavior, reflecting differences in emission sources, atmospheric lifetimes, and chemical transformation pathways.

3.1. Ozone (O₃)

During the pre-fire phase, the average tropospheric O₃ column concentration over the Los Angeles region was approximately 0.121 mol/m². This value increased to ~0.126 mol/m² during the active fire period, representing a ~4.1% increase compared to the pre-fire baseline. Following the fire, in the post-fire period, the average O₃ concentration declined to ~0.123 mol/m², which is ~1.7% higher than pre-fire values, indicating a partial return toward baseline conditions.

This modest and short-lived enhancement of O₃ can be attributed to its role as a secondary pollutant, formed through photochemical reactions involving wildfire-emitted precursors such as NO_x and VOCs [22,23]. However, the formation of O₃ was constrained by low solar intensity in winter, which limits the photolysis of NO₂, a critical step in ozone formation. Furthermore, high concentrations of NO in fresh fire plumes likely led to ozone titration (NO + O₃ → NO₂ + O₂), effectively suppressing ozone accumulation in areas of intense burning [24]. These factors explain the relatively weak O₃ response, despite significant precursor emissions during the fire.

3.2. Nitrogen Dioxide (NO₂)

The mean NO₂ concentration during the pre-fire period was ~0.00015 mol/m². During the wildfire, this value rose sharply to ~0.00020 mol/m², indicating a ~33% increase. In the post-fire period, NO₂ declined to ~0.00017 mol/m², remaining ~13% above pre-fire levels.

The sharp rise in NO₂ reflects its role as a primary pollutant, directly emitted from combustion of nitrogen-containing biomass [25]. The temporal alignment of the NO₂ peak with the fire duration underscores the immediate atmospheric response to wildfire emissions. The partial persistence of elevated NO₂ post-fire may result from lingering pollution trapped near the surface under stable atmospheric conditions or slower vertical dispersion [25,26]. The disparity between the pronounced NO₂ rise and the moderate O₃ response emphasizes the chemically decoupled behavior of primary and secondary pollutants under wintertime wildfire conditions. While NO₂ reflects direct emissions, O₃ formation is chemically dependent on environmental factors that were suboptimal during this period [25,27].

3.3. Carbon Monoxide (CO)

CO exhibited the most substantial increase among the analyzed pollutants. From a pre-fire average of ~0.035 mol/m², CO concentrations increased to ~0.040 mol/m² during the fire, corresponding to a ~14% rise. The maximum daily concentration reached approximately 0.147 mol/m², which is over 320% higher than the pre-fire maximum. Post-fire CO values decreased to ~0.033 mol/m², slightly below the pre-fire mean but still within a range suggesting residual influence.

This high increase is due to CO’s emission as a major byproduct of incomplete combustion of carbon-rich vegetation and urban fuels [28]. Wildfires burning under oxygen-limited conditions release large quantities of CO. Additionally, CO’s long atmospheric lifetime (several weeks) enables it to persist and accumulate in the lower troposphere, especially during winter, when oxidative removal via hydroxyl radicals (OH) is slower due to weaker photochemistry [28,29]. The pronounced and sustained CO enhancement illustrates the efficiency of wildfires in injecting long-lived, health-relevant pollutants into the atmosphere. It also reinforces the importance of CO as a robust tracer for wildfire smoke transport and combustion intensity in satellite-based air quality assessments.

3.3. Aerosol Optical Depth

During the pre-fire phase, the average AOD over the Los Angeles metropolitan region was approximately 0.143, reflecting typical wintertime background aerosol levels under relatively stable meteorological conditions. However, during the active fire period, the mean AOD increased sharply to 0.184, representing a 28.6% enhancement relative to the baseline. This substantial increase corresponds with the massive injection of smoke and fine particulate matter into the troposphere due to intense biomass burning. Wildfires emit large quantities of primary carbonaceous aerosols (e.g., black carbon and organic carbon) and ash particles, which significantly increase atmospheric optical depth [30,31].

Post-fire AOD values averaged 0.163, remaining 14.0% higher than the pre-fire baseline. The persistence of elevated AOD after the fire can be attributed to several factors: (i) slow dispersion of smoke plumes under weak synoptic wind conditions typical of Southern California winters; (ii) and possible long-range transport of aerosols from nearby active fire zones or regional recirculation patterns [32,33]. Moreover, wintertime boundary layer inversions frequently trap pollutants near the surface, exacerbating local aerosol retention [34].

Figure 5. Spatial distribution of mean AOD over the Los Angeles region during the three wildfire phases: (a) pre-fire (b) active fire and (c) post-fire.

Figure 5. Spatial distribution of mean AOD over the Los Angeles region during the three wildfire phases: (a) pre-fire (b) active fire and (c) post-fire.

5. Conclusions

This study assessed the impact of the January 2025 Southern California wildfires on urban air quality over the Los Angeles region using satellite-derived observations of O₃, NO₂, and CO. By analyzing temporal trends and spatial distributions across the pre-fire, during-fire, and post-fire phases, we were able to characterize the pollutant-specific responses to wildfire activity and identify their underlying atmospheric drivers.

The results revealed that CO and NO₂ experienced strong and immediate increases during the active fire period due to their direct emission from biomass combustion. CO concentrations rose by over 14% on average and peaked at values more than 300% above pre-fire levels, while NO₂ increased by approximately 33%, reflecting intense nitrogen emissions from the fire. In contrast, O₃ exhibited a relatively modest increase of ~4%, constrained by wintertime photochemical conditions and likely suppressed further by titration with freshly emitted NO.

Post-fire analysis indicated that while NO₂ and CO gradually declined after the fire, CO exhibited a longer persistence, consistent with its longer atmospheric lifetime and slower removal mechanisms. O₃ returned to near baseline levels, reinforcing its dependence on solar-driven chemistry and secondary formation dynamics.

These findings highlight the value of satellite remote sensing for wildfire air quality monitoring, particularly in data-sparse or rapidly evolving fire scenarios. The differentiated pollutant responses underscore the importance of accounting for both primary and secondary atmospheric processes when assessing the health and environmental consequences of wildfires. This study contributes to ongoing efforts to understand and mitigate wildfire-induced pollution in urban regions under changing climate and land-use conditions.